跳到正文
arXiv cs.AI· Karthik Iyer, Madan Dabbeeru·· 2 小时前AI 评分37

零售销量预测模型对比研究:树模型、基础模型、Transformer 与世界模型在四个公开数据集上的较量

Forecasting Retail Sales at Every Grain: Trees, Foundation Models, Transformers and World Models Compared on Four Public Data Sets

AI 导读

一篇 arXiv 论文在 M5、Dominick's、Breakfast at the Frat 和 Corporacion Favorita 四个公开数据集上,比较 LightGBM、TimesFM 等预训练基础模型、Temporal Fusion Transformer 和世界模型在多粒度零售销量预测中的表现。

正文 · 原文

View PDF HTML (experimental)

Abstract:A retailer makes decisions from the same demand at different grains: store orders by item and day, promotions by item and week, category reviews and supplier terms. This paper asks whether one forecast, built at the finest grain and summed to every coarser grain, can serve them all, and which model does that best. It compares LightGBM, pretrained foundation models (TimesFM, Chronos-Bolt, Timestral), the Temporal Fusion Transformer and a world model that carries each item's state forward under the price and promotion plan, on M5, six Dominick's categories, Breakfast at the Frat and Corporacion Favorita, with the world model and LightGBM tuned by a common search. When to use the average: the average of the tuned world model and LightGBM, summed to every grain, has the lowest error in 22 of 27 grain and data set cases and is never the least accurate on the final held-out periods. When to use the world model: on existing items kept out of training it loses at most 0.040 in scaled error, against up to 0.311 for LightGBM, and on lines with short history it is clearly more accurate on six of nine data sets. When to use LightGBM: on heavily promoted lines, fast sellers and lines with long history (six of nine data sets each) and in price-cut weeks. Inconclusive: lines never promoted and slow lines are close to even, and 52 weeks ahead the average is the most accurate on five of seven data sets, with the world model ahead of LightGBM on two of seven. The findings hold for these data sets and the price and promotion levers; anyone adopting them should rerun every comparison on their own data.
Comments: 38 pages. Version 3: title, abstract and code archive corrected; paper as version 2. Code and data: doi:https://doi.org/10.5281/zenodo.24271672
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2610.25170 [cs.LG]
  (or arXiv:2610.25170v3 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2610.25170

arXiv-issued DOI via DataCite

Submission history

From: Karthik Iyer [view email]
[v1] Sun, 18 Oct 2026 10:50:36 UTC (532 KB)
[v2] Mon, 16 Nov 2026 07:22:12 UTC (469 KB)
[v3] Tue, 17 Nov 2026 09:23:58 UTC (469 KB)

来源:arXiv cs.AI · arxiv.org